Upload
buikien
View
215
Download
1
Embed Size (px)
Citation preview
Face-space: A unifying concept. 1
Face-space: A unifying concept in face recognition research.
Tim Valentine
Goldsmiths, University of London, London, UK
Michael B. Lewis
University of Cardiff, Cardiff , UK
Petter J. Hills
University of Bournemouth, Poole, UK
Running Head: Face-space: A unifying concept.
Word count: 13,656
Corresponding Author: Tim Valentine, Department of Psychology, Goldsmiths,
University of London, New Cross, London SE14 6NW. email: [email protected]
Phone: +44 (0)207 919 7871.
Face-space: A unifying concept. 2
Abstract
The concept of a multi-dimensional psychological space, in which faces can be
represented according to their perceived properties, is fundamental to the modern theorist in
face processing. Yet the idea was not clearly expressed until 1991. The background that led
to Valentine’s (1991a) face-space is explained and its continuing influence on theories of face
processing is discussed. Research that has explored the properties of the face-space and
sought to understand caricature, including facial adaptation paradigms is reviewed. Face-
space as a theoretical framework for understanding the effect of ethnicity and the
development of face recognition is evaluated. Finally two applications of face-space in the
forensic setting are discussed. From initially being presented as a model to explain
distinctiveness, inversion and the effect of ethnicity, face-space has become a central pillar in
many aspects of face processing. It is currently being developed to help us understand
adaptation effects with faces. While being in principle a simple concept, face-space has
shaped, and continues to shape, our understanding of face perception.
Keywords: face; recognition; caricature; adaptation; ethnicity.
Face-space: A unifying concept. 3
Introduction
Development of formal models of human categorization and recognition requires a
stimulus set in which the dimensions or features on which stimuli vary can be controlled.
Artificial faces were a favorite stimulus set used to develop these models in the 1970s and
early 80s (e.g. Goldman & Homa, 1977; Medin & Schaffer, 1978; Reed 1972; Solso &
McCarthy, 1981). The stimulus sets were constructed in a similar manner to the ‘Identikit’
and ‘Photofit’ facial composite systems of the day (see Figure 1 for an example). A similar
approach was also found in studies of cue saliency in face recognition (e.g. Davies, Ellis &
Shepherd, 1977). The assumption, sometimes implicit, was that faces (or concepts) could be
represented as a collection of interchangeable parts.
Figure 1 about here
During this period theoretical models of concept representation were becoming more
sophisticated. Prototype models of concept representation (e.g. Palmer, 1975) were being
challenged by exemplar models that postulated no extraction of a prototype or central
tendency. Exemplar theorists demonstrated that empirical effects, previously interpreted as
evidence of prototype extraction, could be explained by more flexible exemplar models (e.g.
Nosofsky, 1986). But the concept representation literature was becoming increasingly remote
from understanding how we recognize faces in everyday life. Understanding how stimuli like
those shown in Figure 1 can be represented provided little insight into how the relevant
features or dimensions are extracted from real images of faces to enable us to recognize and
categorize real faces (Figure 2).
Figure 2 about here
Face-space: A unifying concept. 4
Ellis (1975) published an influential review that highlighted the lack of theoretical
development in the face processing literature. Responding to this criticism, a literature on the
recognition of familiar (e.g. famous) faces developed, drawing on a theoretical framework
from word recognition, especially Morton’s logogen model (e.g. Morton, 1979). This
approach led to the development of a leadingmodel of familiar face processing (Bruce &
Young, 1986). However, this model had little to say about the visual processing of faces or
recognition of unfamiliar faces. The theory of recognition of familiar faces and of unfamiliar
faces had become separated.
Face-space was motivated by the aim to find a level of explanation, relevant to both
familiar and unfamiliar face processing, which avoided the theoretical cul-de-sac of cue
saliency. The framework was intended to draw on theories of concept representation, while
avoiding the lack of ecological validity of artificial categories of schematic face stimuli. An
important principle was that face-space would capture how the natural variation of real faces
affected face processing.
One of the theoretical contributions that Ellis (1975) reviewed was work on the effect
of inversion on face recognition (Yin 1969). Goldstein and Chance (1980) had suggested that
effects of inversion and ethnicity could both be explained by schema theory. They argued that
as a face schema developed it became more “rigid”: tuned to upright faces and own-ethnicity
faces. Support for the theory came from work showing that the effects of inversion and
ethnicity were less pronounced in children who were assumed to have a less well developed,
and therefore less rigid, face schema (Chance, Turner, & Goldstein, 1982; Goldstein, 1975;
Goldstein & Chance, 1964; Hills, 2014). Schema theory provided an encompassing theory for
face recognition but lacked the specificity required to derive many unambiguous empirical
predictions.
Face-space: A unifying concept. 5
Light, Kyra-Stuart and Hollander (1979) applied schema theory to study of the effect
of the distinctiveness of faces. These authors demonstrated an effect of distinctiveness on
recognition memory for unfamiliar faces. Recognition was more accurate for faces that had
been rated as being more distinctive or unusual, than for faces rated as typical in appearance.
Light et al. interpreted the effect of distinctiveness as evidence of the role of a prototype on
face processing. Influenced by Goldstein and Chance’s application of schema theory and the
work by Leah Light and her colleagues on distinctiveness in recognition memory for
unfamiliar faces, Valentine and Bruce argued that if faces were encoded by reference to a
facial prototype, an effect of distinctiveness should be observed in familiar face processing.
Valentine and Bruce (1986a) found that famous faces rated as being distinctive in appearance
were recognized faster than famous faces rated as being typical, when familiarity was
controlled. Independent effects of distinctiveness and familiarity on the speed of recognizing
personally familiar faces were observed (Valentine & Bruce, 1986b). The effect of
distinctiveness was found to reverse with task demands. Distinctive faces were recognized
faster than typical faces; but took longer than typical faces to be classified as faces when the
contrast category was jumbled faces (Valentine & Bruce, 1986a). These effects of
distinctiveness were explained in terms of faces being encoded by reference to facial
prototype. The final chapter of Valentine (1986) aimed to provide an overarching framework
to conceptualize the effects of distinctiveness, inversion and ethnicity, based upon the
representation of faces by a facial prototype in multi-dimensional similarity space. Valentine
(1991a) was the first publication of this framework. This paper added a version of face-space
in terms of an exemplar model, without an abstracted representation of the central tendency.
It also included empirical tests of predictions derived from the framework.
A Unifying Model
Face-space: A unifying concept. 6
Face-space is a psychological similarity space. Each face is represented by a location
in the space. Faces represented close-by are similar to each other; faces separated by a large
distance are dissimilar. The dimensions of the space represent dimensions on which faces
vary but they are not specified. They may be specific parameters, or global properties. For
example, the height of the head, width of a face, distance between the eyes, age or
masculinity may all be considered potential dimensions of face-space. The number of
dimensions is not specified. Faces are assumed to be normally distributed in each dimension.
Thus faces form a multivariate normal distribution in the space. The central tendency of the
relevant population is defined as the origin for each dimension. Thus the density of faces
(exemplar density) is greatest at the origin of the space. As the distance from the origin
increases, the exemplar density of faces decreases. The faces near the origin are typical in
appearance. They have values close to the central tendency on all dimensions. Distinctive
faces are located further from the origin. The distribution of faces in face-space is illustrated
in Figure 3.
Figure 3 about here
When a face is encoded into face-space there is an error associated with the encoding.
When encoding conditions are difficult, the associated error will be high. Therefore, brief
presentation of faces, presenting faces upside-down or in photographic negative will result in
a relatively high error of encoding. Valentine (1991a) did not make any assumption that
inversion required any specific theoretical interpretation. It has been argued that inversion
selectively disrupts encoding of the configural properties of faces (e.g. Yin, 1969; Diamond
& Carey, 1986). Face-space is agnostic on this issue; it merely treats any manipulation that
reduces face recognition accuracy as increasing encoding error.
Encoding error is likely to result in greater difficulty in recognizing typical faces than
in recognizing distinctive faces (Valentine, 1991a). Typical faces are more densely clustered
Face-space: A unifying concept. 7
in face-space than are distinctive faces, therefore an increase in the error of encoding is more
likely to lead to confusion of facial identify for typical faces than for distinctive faces. There
are fewer face identities encoded near distinctive faces. For a distinctive face, the target
identity is more likely to be the nearest face in face-space even in the presence of a large
encoding error. Valentine (1991a) predicted that presenting faces inverted at test would lead
to a smaller impairment in the accuracy of recognition memory for distinctive faces than for
typical faces. This prediction was confirmed for recognition memory of previously unfamiliar
faces (Experiment 1 and 2). Inversion was also found to slow correct recognition and was
more disruptive to accuracy of recognition of typical famous faces than of distinctive famous
faces (Experiment 3).
An assumption of the face-space framework was that the dimensions of face-space
were selected and scaled to optimize discrimination of the population of faces experienced.
Development of face recognition was assumed to be a process of perceptual learning in which
the dimensions of face-space were tuned to optimize face recognition of the relevant
population. Valentine (1991a) applied face-space to understanding the effect of ethnicity on
face processing. If it is assumed that an observer has encountered faces of only one ethnicity,
with sufficient experience their face-space would be optimized to recognize faces of this
ethnicity. If this observer now started to encounter faces of another ethnicity, faces from a
different population would be encoded in the face-space (the other-ethnicity). Other-ethnicity
faces would be normally distributed on each dimension of face-space but may have a
different central tendency from own-ethnicity faces. Furthermore, some dimensions may not
serve well to distinguish between other-ethnicity faces. But some dimensions that could serve
well to distinguish the other-ethnicity faces may be inappropriately scaled to distinguish the
faces optimally (i.e. the optimal weight required for dimensions may be different between
populations). This situation is illustrated in Figure 4. The other-ethnicity faces form a
Face-space: A unifying concept. 8
relatively dense cluster separate from the central tendency of own-ethnicity faces. In this way
face-space naturally predicts an own-ethnicity bias (OEB1) by which, dependent upon the
observer’s perceptual experience with faces, own-ethnicity faces are likely to better
recognized than faces of a different ethnicity. Valentine and Endo (1992) found that
distinctiveness affected accuracy of recognition memory for previously unfamiliar own-
ethnicity and other-ethnicity faces. Distinctive faces were better recognized than typical faces
in both own- and other-ethnicity populations. The effect of ethnicity on accuracy of face
recognition (Valentine & Endo, 1992, Chiroro & Valentine, 1995) was attributed to the other-
ethnicity faces being more densely clustered in face-space because the dimensions of face-
space were sub-optimally scaled for other-ethnicity faces. With appropriate experience face-
space becomes optimized so that own-ethnicity and other-ethnicity faces are recognized
equally well. However, Chiroro and Valentine (1995) reported two qualifications to this
effect. First, sheer exposure to other-ethnicity faces is not sufficient to learn to recognize the
faces appropriately. It was only when the social environment required participants to learn to
recognize a number of other-ethnicity faces that they showed the ability to do so. Second,
participants who had learnt to recognize another ethnicity efficiently showed a small effect of
recognizing their own-ethnicity less effectively than participants who had never encountered
the other-ethnicity faces. This could have been predicted from the face-space framework,
because the dimensions have been scaled to recognize two different populations requiring
weights on dimensions that may be slightly sub-optimal for both populations. Recognizing
faces from two populations efficiently is a more difficult statistical problem to solve than
recognizing a single population.
Figure 4 about here.
1 It has been common practice in the literature to use the term "race". However the correct term is ‘ethnicity’, because there is only one human sub-species (race) alive on the planet (Homo Sapiens Sapiens). Even if ‘race’ is regarded as acceptable to refer to the major anthropological groups, it is incorrect (as is common in the literature) to use the term ‘race’ to refer to ethnicities such as ‘Hispanic’ who are, of course, Caucasian and therefore the same race as ‘Whites’.
Face-space: A unifying concept. 9
Care needs to be taken interpreting face-space when it is represented in just two
dimensions as it is in Figures 3 and 4. Face-space was always envisaged as a
multidimensional space with many more than two dimensions. Burton and Vokey (1998)
describe the potential dangers of using a two dimensional representation of what should be a
multi-dimensional space. They argue that, contrary to the intuition derived from a two-
dimensional space, if a space with 1000 dimensions was populated with 1000 normally
distributed exemplars, all of the exemplars would be a similar distance from the origin of the
space; approximately 1000 times the standard deviation of the normal distribution. Hence, in
a high-dimensional face-space there would be few highly typical faces close to the origin.
This point was previously made by Craw (1995). As Burton and Vokey acknowledge it
remains the case that, even in a very high dimensional face-space, the origin of the space is
the point of maximum exemplar density and therefore the predictions of the effects of
distinctiveness in recognition and classification tasks are valid.
A multi-dimensional space differs from the two dimensional illustration in the
expected distribution of distinctiveness (typicality) ratings. The two dimensional figure leads
to the expectation that many faces would be rated as highly typical with progressively fewer
faces given higher ratings of distinctiveness. Burton and Vokey (1998) observed that, instead,
typicality ratings of faces are normally distributed. Most faces are judged to have moderate
levels of typicality, with few rated as highly typical, or highly distinctive. Burton and Vokey
demonstrated that this distribution is predicted by a multidimensional normal distribution, as
assumed in the face-space model. The point Burton and Vokey made was that it can be
misleading to generalize from simple two dimensional representations to high dimensional
spaces. Mathematic analysis, rather than intuition, is required to evaluate the predictions of
such a model.
Face-space: A unifying concept. 10
Although Burton and Vokey (1998) did not extend their analysis to consider the
attractiveness of faces, their analysis does explain a paradox in the literature. Morphing faces
to produce an average facial appearance produces a face that is strikingly attractive. This
effect was first observed by A. L Austin (Galton, 1878, [see Valentine, Darling & Donnelly,
2004]) and more recently has been demonstrated formally (e.g. Langlois & Roggman, 1990;
Perret, May and Yoshikawa, 1994). This work suggests that typical faces are highly
attractive. The paradox is this: If typical faces are common in the population, why are highly
attractive faces rare? Burton and Vokey’s analysis provides the answer: very typical faces are
rare; therefore highly attractive faces are rare. It is rare for faces that can vary on many
dimensions to be average on all of them.
The original formulation of face-space did not specify the nature of its dimensions. It
was always considered that the dimensions might be holistic (e.g. age, gender or face-shape).
One way to operationalise face-space is to equate the dimensions of face-space with the
components derived from principal component analysis of facial images or eigenfaces (Turk
and Pentland, 1991). The concept of eigenfaces was developed by computer scientists as a
method to compress the information in a set of faces. This conceptualization of face-space
has been widely used by computer scientists and, amongst other applications, is used to
generate synthetic composite faces. The approach is reviewed below under the section on
forensic applications.
In summary, the face-space framework described by Valentine (1991a) unified the
accounts of the effects of distinctiveness, inversion and ethnicity on face recognition.
Valentine (1991b) extended the approach to include an account of caricature. The approach
was to provide a framework which, although underspecified, could be applied to
understanding variation in a real population of faces. Use of artificial stimulus sets was
rejected as an appropriate tool to understand face recognition in the real world.
Face-space: A unifying concept. 11
Norm-based coding vs. Exemplar model
Valentine (1991a) originally suggested two different models within the face space
framework. The first was one in which faces are encoded relative to a specific prototypical
face also known as a norm face. In this norm-based face-space, faces are coded relative to
this central face. The stored representation can be seen akin to an angle in which the direction
and the magnitude are required to define the location of a face within this space. The
distinctiveness of a face is represented by the length of this vector whereas the direction
defines the identity.
The alternative model of face-space offered in Valentine (1991a) was an exemplar-
based version. In an exemplar-based face-space, the faces are represented in the space
without specific reference to any central prototype. The distance between face representations
provide the measure of their similarity and it is the distribution of faces within the space that
leads to the distinctiveness effects described above. Distinctive exemplars will be in areas of
low density of other exemplars as a consequence of the normally distributed pattern of faces
that one sees and knows. Typical faces, on the other hand, will be located near the centre of
the distribution and thus there will be many similar face representations with which to
confuse a particular exemplar.
This distinction between norm-based and exemplar-based versions of face-space
reflected wider debate on the nature of memory. Exemplar-based models of memory were
developed (e.g., Medin & Schaffer, 1978; Nosofsky, 1986;1988; 1991) as an alternative
account of memory to category knowledge based on the extraction of prototypes (e.g.,
Goldstein & Chance, 1980; Knowlton & Squire, 1993; Palmer, 1975; Reed 1972). There has
been a great deal of research that has been conducted on the domain of face perception that
speaks to the differences between these two models of face-space, included research on the
Face-space: A unifying concept. 12
own-ethnicity bias, caricature recognition and more recently facial adaptation effects. The
contribution of each of these topics to our better understanding of face-space will be reviewed
in turn, but first it is worth looking at the formulations of these two differing models in more
detail.
Similarity metrics
To formalize the differences between norm-based face-space and exemplar-based
face-space it is necessary to consider the similarity metrics that define them. A basic
assumption for all metrics is that faces that are similar are encoded close together in the
space, and therefore are confusable. While all versions of face-space suggest that faces are
encoded in a multi-dimensional space, the properties of this space can differ. The most
important property is how similarity of two faces maps onto distance in the face-space. This
is the similarity metric.
As a working hypothesis, Valentine (1991a) defines the similarity metric for the
exemplar-based model as the simple Euclidean distance between the exemplars. Recognition
takes place if a target’s representation is sufficiently similar to an encoded representation of a
known exemplar but sufficiently dissimilar from the next most similar encoded exemplar. A
development of this recognition decision based on an exemplar-based similarity metric was
employed in the computational implementation by Lewis (2004) called face-space-R. In this
model, a distribution of ‘faces’ was generated such that they were normally distributed on
each dimension of face-space (i.e. a multi-dimensional normal distribution) and tested in a
variety of tasks. The similarity metric employed was such that two identical faces had
maximal similarity but similarity between faces decreased as Gaussian decay function with
distance in the space. Lewis demonstrated that findings concerning distinctiveness, ethnicity
and caricatures could be accounted for using this similarity metric.
Face-space: A unifying concept. 13
One consequence of this type of exemplar-based similarity metric is that if two faces
differ by the same distance in the space they will be equally similar regardless of whether
they are typical or distinctive. There is now some evidence that this is not the case. Ross,
Hancock and Lewis (2010) generated sets of stimuli where the same physical change was
either applied such that the modified face lay on a radial line from the average face to the
location of the original face in face-space, or the new location was oblique to a line between
the exemplar and the average face. A discrimination task found that changes along the radial
line from average (norm) face were harder to detect than changes that were oblique to that
line.
The similarity metric for the norm-based model was not clearly defined in Valentine
(1991a) except for the suggestion that it was based on vector similarity. Some authors have
taken this to mean the dot product of the vectors. However, the dot product would predict that
two vectors, representing different faces, would appear more similar to each other as one of
them increased in magnitude (e.g. became more distinctive by being caricatured). Byatt and
Rhodes (1998) proposed a similarity metric defined by the cosine of the angle between the
vectors’ representations of two faces (relative to a norm face) divided by the simple distance
between the two faces. The benefit of this metric was that faces that were on the same radial
axis were more similar to each other than those that where equidistant but were not on the
same radial axis. The metric was also able to distinguish between two faces that lay on the
same radial axis but were still different distances from the average face.
The question as to the correct metric for face-space cuts right to the definition of face-
space itself. If the metric is not calculated relative to a norm face then the face-space is not
norm-based. There remains no consensus on the correct interpretation of the similarity metric.
As such, the question as to the role of a norm face in face recognition remains an open one.
Caricatures
Face-space: A unifying concept. 14
The recognition of caricatures has been influential, but controversial, in revealing the
nature of face-space. Artists’ portrayals of caricatures are better recognized than veridical
images (Perkins, 1975). A similar finding was found with computer-generated caricatures
(Benson & Perrett, 1991). Such computer-generated caricatures can be produced by the
following process. The location of many facial landmarks, which define the shape of the
face’s appearance (e.g. corners of eyes, outline of the nose etc.), are recorded for many faces
from a homogeneous population (e.g. male White faces). The locations are averaged to define
a ‘norm’ or ‘prototype’ face. A computer-generated caricature of an individual face can then
be generated by exaggerating all the differences in the location of the landmarks between the
individual face and the average face by a fixed proportion (e.g. 30%, 50%, see Figure 5). The
proportion of the exaggeration defines the extent of the caricature. The visual texture can then
be scaled and re-mapped to fit the new facial shape. This process exaggerated distinctive
features. For example, an atypically large nose becomes even larger in a computer-generated
caricature. Anti-caricatures (in which differences from the average were reduced) were also
constructed.
Figure 5 about here
The fact that caricatures are recognized more accurately than veridical images is most
easily explained by norm-based versions of the face-space. A caricature will have a
representation that has the same angle from the prototypical face but will have a longer
vector. This longer vector has been argued to be the parameter that provides the improved
recognition of caricatures over veridical faces; in effect the caricature is a super-stimulus of
the facial identity.
The exemplar-based version of the face-space, however, is not silent on the topic of
caricatures. This is because, although exaggerating a face away from the average makes the
face more unlike its target representation, it also makes it less like any competitor
Face-space: A unifying concept. 15
representations as well. Lewis and Johnston (1999) had shown that an advantage for
recognition was found for images that were exaggerated away from other similar known
faces. This fact was used in the face-space-R model (Lewis, 2004) to demonstrate how an
exemplar-based face-space predicts better recognition for a caricature face over a veridical
face. The model was also able to make estimates of the degree of caricature that would lead
to optimal recognition. Through modelling the caricature data, Lewis was able to make an
estimate for the number of dimensions that we may use in a face-space. The estimate was
between 15 and 22 dimensions.
The fact that both the norm-based model and the exemplar-based model can predict a
recognition advantage for caricatures has recently become an interesting issue, as the
existence of a caricature advantage has been drawn into question. The studies that do show a
strong caricature advantage tend to use impoverished stimuli either because they are line
drawings (e.g., Rhodes, Brennan & Carey, 1987) or because they are presented briefly (Lee
& Perrett, 1997). Some studies only show an advantage for caricatures over anti-caricatures
(Lewis & Johnson, 1998). More recent studies demonstrate no advantage for the caricature
(Kaufmann & Schweinberger, 2008) or even an advantage for the anti-caricature (Allen,
Brady & Tredoux, 2009). Indeed, Hancock and Little (2011) suggest that the reason for the
caricature advantage observed in earlier studies was at least partly due to adaptation effects as
a result of the way in which the stimuli were presented. Exactly how these adaptation effects
work and what they tell us about face-space is explored further below. The situation is that
there are two models that each predict a caricature advantage but there is debate over whether
the effect on recognition is real or an artifact. Further research is required to resolve this.
Facial adaptation.
Facial adaptation effects have demonstrated how the face-space is a flexible concept
and representations can be distorted within it. Adaptation is a recalibrating process in which
Face-space: A unifying concept. 16
the perceptual system is altered following constant stimulation of a particular stimulus
characteristic (Blakemore, Nachmias, & Sutton, 1970). One of the first demonstrations of
face adaptation was shown by Lewis and Ellis (2000), although they used the term satiation
rather than adaptation. They showed that the time required to recognize a face increased
when 30 different views of that face had been presented immediately before the test
(compared with just 3 different views). As well as slowing recognition, adaptation also
causes contrastive after-effects, such that adaptation to a center-compressed facial image
causes the perception of an unaltered image to appear center-expanded (Rhodes & Jeffery,
2006; Webster & MacLin, 1999; see Figure 6). This is the typical face distortion after-effect
(FDAE). Contrastive facial after-effects have also been observed for judgments of
attractiveness (Rhodes, Jeffery, Watson, Clifford, & Nakayama, 2003), personality
(Buckingham et al., 2006; Wincenciak, Dzhelyova, Perrett, & Barraclough, 2013), emotion
and gender (Webster, Kaping, Mizokami, & Duhamel, 2004) and identity (Leopold, O’Toole,
Vetter, & Blanz, 2001; Leopold, Rhodes, Müller, & Jeffery, 2005).
Figure 6 about here.
Face after-effects transfer across face identities (even to the perceivers' own face;
Webster & MacLin, 1999), from an adaptor of one size to test stimuli of a different size
(Zhao & Chubb, 2001), across different parts of the retina (Hurlbert, 2001; Anderson &
Wilson, 2005) and partially across viewpoints (Jeffery, Rhodes, & Bussey, 2006; Pourtois,
Schwartz, Seghier, Lazeyras, & Vuilleumier, 2005; Ryu & Chaudhuri, 2006), yet visual
similarity between the adaptor and test is a critical variable in the magnitude of the FDAEs
(Yamashita, Hardy, De Valois, & Webster, 2005) at least for unfamiliar faces (Hills & Lewis,
2012). For familiar faces, there is greater transference across size and viewpoint (Carbon &
Leder, 2005; Jiang, Blanz, & O’Toole, 2006), indicating that face after-effects involve
higher-level perceptual processing than observed in other after-effects.
Face-space: A unifying concept. 17
Not all kinds of distortions can cause the FDAE, however. Robbins, McKone, and
Edwards (2007) demonstrated that after-effects were observed when participants were
adapted to a “natural” facial configuration (eyes aligned) but not when adapted to “unnatural”
facial configurations (eyes not aligned). The after-effect transferred from the adaptor identity
to other faces. This indicates that adaptation techniques may be useful in revealing the nature
of the dimensions of face-space.
Often considered similar to FDAEs are face identity after-effects (FIAEs; Strobach &
Carbon, 2013; Webster & MacLeod, 2011), whereby the perceived identity of a face is
altered after adaptation to a particular identity. Leopold et al. (2001) morphed together 200
faces to produce a prototype face. This was assumed to be the centre of the face-space. Each
unique face identity could be measured in terms of Euclidean distances from the prototype
face. After adaptation to an anti-face (a projection through the origin of face-space in the
opposite direction from the face-identity), the identification threshold (the required identity
strength to perceive the face identity) was lowered by 12.5% suggesting it was easier to
perceive the identity following adaptation (see Figure 7). These effects are considerably
weaker if the adapt and test face continuum do not pass near the norm (Zhao, Hancock, &
Bednar, 2008). Nevertheless, since they are still present, these results demonstrate that
adaptation to one particular face alters the entire face-space (Benton & Burgess, 2008).
Figure 7 about here.
These facial adaptation effects are easily explained in terms of the face-space using
Clifford, Wenderoth, and Spehar's (2000) notion that the sensory system dynamically maps
environmental attributes onto patterns of fixed neuronal responses. This mapping is changed
when the structure of the environment is altered. Implicit within this model is that neuronal
populations have tuning curves for particular stimuli characteristics which may correspond to
dimensions of face-space. The width of tuning curves represents the population’s response
Face-space: A unifying concept. 18
bandwidths, whereas the peak represents the preferred stimulus property. The perceived
response is given by the weighted vector average of the units responding to the stimulus,
based on distribution-shift theory (c.f. Mather, 1980). The purpose of adaptation is to recentre
the perceptual space nearer to the adaptor stimulus (Hurlbert, 2001; Webster & Macleod,
2011). Evidence for the renormalization process of adaptation stems from the fact that it is
not possible to adapt to a prototype face (Webster & MacLin, 1999). This renormalization
process leads to apparently permanent changes in the face-space (Carbon & Diyte, 2012)
suggesting a role for adaptation is face learning and the creation of the face prototype in face-
space. For example, consistently seeing faces with a wide nose alters the face prototype, such
that wide noses are considered typical and narrow noses distinctive.
Researchers have also used adaptation to conceptualize the neural representation of
the dimensions of face-space. Typically, two-pool models have been employed to account for
what appears to be opponent processing (e.g. Over, 1971) and have been successfully
implemented in modeling the FIAE (Ross, Deroche, & Pameri, 2014). At one end of each
dimension there is a neural population for the extreme of a particular feature and at the other
end of that same dimension, there is a neural population for the opposite extreme of the same
feature (e.g. Robbins et al., 2007, see Figure 8). Equal activation of both pools of neurons
signals a neutral point on that dimension (i.e. the norm). The relative firing of each pool
determines the size of the feature seen. Coding is therefore relative to norm. Based on this
model, after-effects increase monotonically with increasing adaptor extremity (at least up to
an ecologically valid range of values). This is due to more extreme adaptors activating their
preferred channel more strongly (Pond, Kloth, McKone, Jeffery, Irons, & Rhodes, 2013).
Consistently, many face after-effects are found to be larger for strong (e.g., extremely large or
unusual) than weak adaptors indicating norm-based coding for many facial attributes (Burton,
Jeffery, Skinner, Benton, & Rhodes, 2013). Concurrently, evidence from fMRI adaptation
Face-space: A unifying concept. 19
and single-cell recording studies indicate that face-selective neurons are tuned to encode the
distinctiveness of individual faces relative to the prototype face (Leopold, Bondar, & Giese,
2006; Loffler, Yourganov, Wilkinson, & Wilson, 2005).
Figure 8 about here
While this two-pool account is appealing, there are caveats with it. There is evidence
to suggest that some of the dimensions along which faces are thought to vary are not
orthogonal: the effects of adaptation on one dimension may depend on the levels of other
dimensions (Jaquet & Rhodes, 2008). Studies have shown that when participants are adapted
to opposing pairings of facial characteristics category-contingent after-effects are typically
produced (e.g. adaptation to expanded eye spacing in White faces and constricted eye spacing
in Black faces at the same time causes contrasting after-effects in White and Black faces).
Aftereffects are not therefore based on the simple translation of experienced face-space
relative to physical face-space. Therefore, norms for different face categories can be
established depending on the context. Thus, there may be separate face-spaces for different
categories of faces (Little, DeBruine, Jones, & Waitt, 2008). Furthermore, FIAEs caused by
familiar faces are only observed if the familiar face is recognized (Laurence & Hole, 2012),
suggesting the involvement of higher-level recognition-based systems at play in the face-
space.
Adaptation effects can also be explained within a multichannel or exemplar-based
version of face-space where a facial attribute is represented by activation in many pools of
neurons. Each pool is tuned to a narrow range of distinct values. An exemplar of a feature is
represented by the summation of activation in several of these pools. These pools are
maximally tuned to naturally occurring values. Therefore, extreme adaptors (outside the
range that would be expected in the real-world) would cause less after-effect than less
extreme adaptors as they have less impact on the channels in the less extreme range. These
Face-space: A unifying concept. 20
results have recently been modeled (Ross et al., 2014) indicating that an exemplar-based
model of face-space can account for after-effects.
Adaptation can also speak to how inverted faces may be represented in the face-space.
In Valentine's (1991a) original model, inversion leads to increased error in encoding.
Research exploring after-effects and orientation show that after-effects occur for upright and
inverted faces, but only if the orientation of the adaptation face was matched with the
orientation of the test faces (Webster & MacLin, 1999). This suggests that inverted faces are
not processed in the same face-space as upright faces, or that the dimensions used to code
faces are orientation specific. Such category-contingent after-effects further indicate the
possibility of their being many face-spaces. This begs the question of how many face-spaces
there are and how they may be structured: It is possible that there is a hierarchical structure
whereby there is a face-space for faces overall with further spaces for specific categories
(such as male/female), or that multiple spaces exist and motivation dictates which space is
used. These are clear avenues for further research. Whether face after-effects are best
accounted for in a norm-based or exemplar-based face-space, there is no doubt that the
original model of face-space has proven to be a useful framework with which to explore
after-effects. Importantly, these high-level face-specific after-effects and presumably neural
response in face-space, correlate with face recognition accuracy suggesting the more
responsive an individuals' face-space the better they are at face recognition (Dennett,
McKone, Edwards, & Susilo, 2011).
Own-ethnicity bias
Valentine's (1991a) explanation of the OEB is based on the observed fact that other-
ethnicity faces do not vary consistently along the same dimensions as own-ethnicity faces,
due to the physiognomic differences (for example between Black and White faces, Ellis,
Deregowski, & Shepherd, 1975). Indeed, Papesh and Goldinger (2010) have shown, using a
Face-space: A unifying concept. 21
multidimensional scaling (MDS) approach, that participants' perceptual space grouped highly
controlled Black and White faces separately, with other-ethnicity faces grouped more densely
in the MDS space. This grouping was based on structural properties of the faces rather than
skin tone. These results are entirely consistent with face-space and indicate that the
dimensions used for encoding and recognition of own-ethnicity faces are diagnostic and
appropriate, but are unlikely to be as diagnostic for the processing of other-ethnicity faces
(Hills & Lewis, 2011). Based on this idea, a number of authors have attempted to reduce the
OEB by training participants to focus on features that differentiate other-ethnicity faces
(Slone, Brigham, & Meissner, 2000). Hills and Lewis (2006) trained White participants to
use the features typically described by Black participants when recognizing faces. This led to
a reduction in the OEB amongst White participants because they looked at the diagnostic
features (Hills & Pake, 2013).
While this explanation of the OEB is parsimonious, there are other recognition biases
that are harder to explain within the face-space framework. The own-age (Anastasi &
Rhodes, 2005, 2006), own-gender (Wright & Sladden, 2003), and own-university biases
(Bernstein, Young, & Hugenberg, 2007) are not based on extensive experience: participants
are likely to have roughly equivalent experience of own and other-gender faces; participants
are less able to recognize faces younger than their own age, even though they were once
young (Hills,2012). And finally, own-university biases are not based on physical differences
between groups of faces. These biases may be explained in terms of motivation to process
own-group faces deeply (Sporer, 2001). Nevertheless, expertise is required to differentiate
and process faces deeply.
To explain these biases within the face-space framework, one can assume that
dimensions that are diagnostic for differentiating between other-age faces can become more
heavily–weighted and therefore more salient, depending on the age of the participant. Indeed,
Face-space: A unifying concept. 22
recent daily-life contact influences the magnitude of the own-age bias (e.g., Wiese, Komes, &
Schweinberger, 2012). This perspective indicates that the face space adapts to the present
living conditions. A useful analogy is to think of face-space around own-age and own-gender
faces being stretched, to make these faces more dissimilar to each other and therefore easier
to distinguish (cf. Nosofsky, 1986), akin to perceptual warping (Kuhl, Williams, Lacerda,
Stevens, & Lindblom, 1992). This inherently adds a flexibility element to face-space.
Dimensions do not have to be used equivalently in every situation, but can be weighted
according to the task and motivation. A lack of motivation to process other-group faces may
also lead to some dimensions being weighted more than others, leading to increased error
during the encoding. This indicates an unanswered question regarding how own- and other-
age faces are distributed within the face-space and how this changes with age.
Development of face–space
Given the importance of face-space for explaining how adults encode faces, it is
important to apply this model to the development of face recognition. Face recognition
abilities improve through childhood (e.g. Blaney & Winograd, 1978; Ellis & Flin, 1990; List,
1986). This improvement may follow a linear pattern (Feinman & Entwisle, 1976; Hills,
2014) and is associated with better memory for faces in children (Dempster, 1981; Hills
2012) and faster responding (Johnston & Ellis, 1995). Children's face recognition is inexpert.
Children show a smaller effect of inversion on face recognition than do adults, when tested
using appropriate methods (de Heering, Rossion, & Maurer, 2012; Hills, 2014; Rose,
Jankowski, & Feldman, 2002). Children recognize typical and distinctive faces equally well
(Johnston & Ellis, 1995). Some effects are simply observed differently. Children can be
adapted to unnatural facial configurations that adults cannot (Hills, Holland, & Lewis, 2010).
These effects can be interpreted within the face-space framework. Evidence suggests
that the face-space or prototype face becomes increasingly more refined with age enabling
Face-space: A unifying concept. 23
improved differentiation of categories of faces to be made (Short, Hatry, & Mondloch, 2011).
This refinement may stem from children’s use of the dimensions of face-space, the
distribution of faces within face-space, or the use of the prototype. These three possibilities
relate to the number of dimensions of face-space, the coding accuracy along the dimensions,
and the coding of the prototype.
Johnston and Ellis (1995) presented the uniform model of face-space, in which the
number of dimensions of face-space is constant throughout development. The distribution of
faces within face-space changes, starting with an empty face-space. As faces are encountered,
they are stored in the space. This creates crowding in the centre of the face-space and leads to
typicality effects and own-group biases that will not be present in younger children.
An alternative position is that children may use fewer dimensions than adults
(Nishimura, Maurer, & Gao, 2009). These may be “less appropriate” dimensions than those
of adults (Johnston & Ellis, 1995 pp. 463), suggesting that dimensions will be added to the
face-space during development (Lewis, 2004; Valentine, 1991a) when two faces cannot be
readily discriminated with the existing dimensions. Perceptual learning (e.g. McLaren, 1997;
Mundy, Honey, & Dwyer, 2007) indicates that when two similar stimuli are presented,
participants will attend to the unique differences between the stimuli and inhibit the
similarities. Therefore, a dimension that distinguishes between the two faces may be added to
the face-space. This process will continue until most faces are recognizable. Thus, experience
of faces dictates what dimensions will be added to face-space. With this position, there is no
specificity about when dimensions will no longer be added to the space.
A third position is that children may use different or a larger number of dimensions
than adults (Hills et al., 2010). Perceptual narrowing effects (Nelson, 2001; Macchi-Cassia,
Kuefner, Westerlund, & Nelson, 2006), by which younger infants are better able to
discriminate between pairs of monkey faces and phonemes of non-native languages than
Face-space: A unifying concept. 24
older infants (Cheour, et al., 1998; Kuhl, et al., 1992; Lewkowicz & Ghazanfar, 2006;
Pascalis, de Haan, & Nelson, 2002; Scott, Shannon, & Nelson, 2005, 2006), indicate that with
development children learn to orient their perceptual attention to the most relevant
characteristics (Saarinen & Levi, 1995). Similar perceptual narrowing has been shown to
occur in other realms of expertise such as chess (e.g. Gobet & Simon, 1996a, b) and is based
upon the ability to form category boundaries. Category boundaries are associated with face
expertise (Angeli, Davidoff & Valentine, 2008; Balas, 2012). Therefore, it is possible that
experience narrows the number of dimensions used to recognize faces ensuring that only the
most appropriate dimensions remain for frequently encountered faces. Potentially, children
may rely more on featural dimensions, as opposed to configural dimensions, than adults and
this may explain smaller face-inversion effects. With age, the face-space becomes more
refined and specific to the most frequently encountered faces (upright, own-ethnicity faces)
and this is evidenced by apparent coding changes with age. There is neurological evidence
for such a possibility: the number of axons in the visual cortex decreases with age leading to
more restrictions in the stimuli that lead to neural responses due to neuronal pruning (Johnson
& Vecera, 1996).
A final position is that children may not be able to code faces precisely along the
same dimensions as adults (Mondloch & Thomson, 2008). Potentially, the neural response is
less distinct for a particular feature. This may lead to a less stable face norm (Nishimura et
al., 2008). The less stable norm may reflect the change from concrete to formal operational
thought in Piagetian terms (Piaget, 1952). Whereas concrete operational thought involves
simple flexible schemas, formal operational thought involves the use of many more complex
schemas (c.f. K. Nelson, 1981; 1996). Therefore formal operational thought could lead to the
development of a robust, ‘rigid’ prototype derived from the most frequently encountered
faces.
Face-space: A unifying concept. 25
All of these versions of face-space make similar predictions regarding the face
processing abilities of children: Children show reduced ability to recognize faces, smaller
face inversion, distinctiveness, and own-group bias effects. Distinguishing between these
models will prove a challenge for future research. The number and use of dimensions of face-
space in childhood remains an open question for future research. A further question relates to
how the face-space develops in older adulthood. There is evidence to suggest that the face-
space remains sensitive to the faces encountered in daily-life even into older adulthood
(Wiese et al., 2012) despite changes in overall memory performance (Komes, Schweinberger,
& Wiese, 2014). This leads to another line of further enquiry based on face-space: how does
it change in latter adulthood when perception and memory abilities change?
Forensic applications of face-space.
We conclude this review by considering two practical applications of face-space in
eyewitness identification. The first application directly uses an image-derived face-space to
represent, and manipulate facial appearance. The second application is the design of an
effective but fair technique to test a witness’ ability to identify a suspect with a distinctive
feature. To evaluate lineups, memory for facial appearance and separately for an additional
distinctive feature is modeled using a hybrid model in which faces are represented both by
their location in a multi-dimensional face-space and in terms of shared and unshared discrete
features.
Computer-generated facial composites.
If an eyewitness is available to an investigation of a serious crime, but there are few
other lines of enquiry, the police may ask the witness for help constructing a likeness of the
offender. Construction of a likeness– known as a facial composite- would normally only be
attempted with a witness who had a good opportunity to view the offender’s face. The
composite can be circulated to police officers or the media, in the hope that somebody who
Face-space: A unifying concept. 26
knows the offender will provide a possible identification and so open a new line of enquiry.
In the past facial composites were constructed, under the guidance of the witness, using
systems that physically swapped facial features (e.g. eyes, nose, mouth) until an acceptable
likeness was achieved (e.g. Photofit, Identikit). These systems performed poorly and the
composites produced were seldom recognized. The requirement to compare and select face
parts out of the context of a whole faces is a difficult task. Human face recognition is strongly
influenced by the subtle configural relationships between facial features. See Davies and
Valentine (2007) for a review and evaluation of facial composite systems
A new generation of composite systems have been developed which draw on the
representation of faces within face-space to ‘evolve’ a facial appearance, using a genetic
algorithm to search face-space under the guidance of the witness for a suitable likeness. The
major difference is that the new systems only ever require a witness to compare whole facial
images. In this way the new ‘holistic’ systems exploit more effectively the natural style of
human face processing.
A facial image similarity space, similar to the concept of face-space, can be
constructed for a large set of standardized facial images by using a statistical method –
Principal Component Analyisis (PCA) - to extract a set of orthogonal factors which serve as
the dimensions of the space. These principal components, or ‘eigenfaces’, are extracted in the
order in which they capture the variance in set of images (e.g. Turk & Pentland, 1991).
Eigenfaces can be thought of as representing the dimensions of face-space. Once the
eigenfaces are specified, any facial appearance can be represented by a set of weights for
each component. The facial appearance can be reconstructed by combining the eigenfaces in
the appropriate proportions specified by the weights. Any (artificial) facial appearance, can
be constructed by a novel combination of weights, within the constraints of the variation of
the original population of faces used to construct the space.
Face-space: A unifying concept. 27
Eigenfaces capture variance across the entire image. Therefore they are holistic in
nature and do not break faces down into component parts. Visualizing a principal component
may be interpretable, for example, as representing gender (e.g. O’Toole, Abdi, Deffenbacher
& Valentin, 1995) but most components are not interpretable. The face-space generated by
principal component analysis of facial images displays an important property: faces
represented close in the space are perceived as similar (Tredoux, 2002). Holistic systems for
facial composite construction are now in common use by police forces. EFIT V and Evo-FIT
are the most widely used systems in the UK. EFIT V is also used in the USA, Canada,
Caribbean, and South America. See Frowd (in press) for a recent review of psychological
research on holistic facial composite systems.
The construction of a facial composite by a witness begins by the generation of a
random set of (artificial) facial images within the PCA space. The witness then selects the
image or images that are most similar to the appearance of the culprit. In the initial set there
will be a wide range of facial appearances and none are likely to closely resemble the culprit.
The selection made by the witness is then used to ‘breed’ a new set of images introducing
mutations around the ‘parent’ face or faces. The process is repeated iteratively, with each
successive ‘generation’ becoming more similar to the culprit and to each other. The process
continues until the witness cannot choose because all of the faces resemble the culprit equally
well, or it becomes clear that the search of face-space has failed to converge on the desired
appearance.
Sometimes a witness says that the offender looked more masculine, or younger than
the current composite image. This kind of manipulation would be impossible in the earlier
feature-based systems, but can be easily implemented in a holistic system by identifying the
direction in face-space that relates to the perception of these characteristics. Solomon, Gibson
and Maylin (2012) described a procedure to identify the relevant dimensions using EFIT V.
Face-space: A unifying concept. 28
Participants are asked to classify a set of faces into binary categories (e.g. male or female), or
rank order faces on a relevant dimension (e.g. age). In the latter case, a median split was then
used to create two categories. A prototype for each category is then derived from the category
members. The direction of the required manipulation in face-space (e.g. to increase
masculinity) is given by the difference vector from the female prototype to the male
prototype. A slider can be provided in the software interface, which applies a manipulation
along this dimension. The perceptual effect is to apply a global transformation that make
faces look more masculine or feminine depending on the direction of travel. This approach
can be applied to any characteristic on which faces can be ordered or categorized.
In summary, conceptualizing the population of faces as represented in a similarity
space (face-space), and implementing that concept as an image-space for a carefully
standardized set of images, has allowed computer scientists to make a radical change in the
construction of facial likenesses by witnesses. Commercial software has given the police a
powerful new tool to identify offenders.
Designing lineups to identify suspects with distinguishing features.
If a witness sees a perpetrator with a distinguishing facial mark or feature (e.g., a scar,
tattoo or piercing) it is very likely that the witness will describe the feature in the description
they give to the police; and quite possible the witness can recall little else about the
perpetrator’s facial appearance. Subsequently, a suspect may be arrested principally because
they have a distinctive feature that fits the witness’ description. If the suspect disputes
identification, the police in England and Wales are required to construct a lineup to test the
ability of a witness to identify the suspect. The question arises of how a fair lineup can be
constructed for a suspect with a distinguishing facial feature. If only the suspect has a tattoo
or piercing, it will be clear to the witness which lineup member is the suspect and the
procedure would be unfair to an innocent suspect.
Face-space: A unifying concept. 29
Currently the police have two options: to conceal the distinguishing feature, or to
replicate the feature on all lineup members. In England and Wales the police almost always
conceal the mark, because concealment can be automated and applied to moving images used
in the standard video identification procedures. Replication has to be applied by hand and can
only be applied to still images. The disadvantage of concealing a distinguishing feature is that
it changes the appearance of the suspect’s face. A distinctive feature is a salient cue to
recognizing the face of a perpetrator (Winograd, 1981). By the encoding specificity principle,
if the suspect is guilty, recognition memory performance will be determined by the overlap of
cues present at encoding and test (Tulving & Thomson, 1973). Therefore, a better strategy is
to replicate the feature on the foils in the lineup, leaving the appearance of the suspect
unchanged. If replication is used, current procedure is to replicate the identical feature on the
faces of all foils.
The hybrid-similarity model of recognition memory (Nosofsky & Zaki, 2003) was
applied by Zarkadi, Wade, and Stewart (2009) to model eyewitness performance for
‘concealment’ and ‘replication’ lineups. In the hybrid-similarity model, similarity between
two faces is a combination of the distance between the faces in a multidimensional similarity
space or face-space (Nosofsky, 1986), and the number of discrete features that are shared
and unshared (Tversky, 1977). The simulation predicted that replication would yield more
culprit identifications from culprit-present (CP) lineups without increasing mistaken
identifications of a foil from culprit-absent (CA) lineups. (A culprit-absent lineup in shown to
a witness when the suspect is innocent – the real culprit is not in the lineup.) The predictions
derived from the hybrid-similarity model were confirmed experimentally by Zarkadi et al.
(2009). Perpetrators with a distinguishing mark (a bruise, a mole, a piercing, a moustache, a
scar, or a tattoo) were more likely to be identified from a six-person simultaneous photograph
lineup when the mark was replicated on all lineup members, then when the distinguishing
Face-space: A unifying concept. 30
feature was removed from the culprit (concealment). Whether the distinguishing feature was
replicated or concealed did not affect the number of foil identifications made from a culprit-
absent lineup.
Following police practice, Zarkadi et al. (2009) applied replication by exactly
replicating the culprit’s distinguishing feature on all foils. This strategy means that the
witness cannot use their knowledge of the distinguishing feature to recognize the perpetrator,
if present in the lineup. Theory of memory suggests that both concealment and replication
strategies are sub-optimal. Valentine, Hughes and Munro (2009) argued that including some
variation in the replication of the distinguishing feature on the faces of foils would increase
the probability of identifying a guilty suspect without increasing the risk of mistaken
identification for an innocent suspect. The distinguishing feature should be replicated with
variation within the constraints of the description of the feature given by the witness. For
example, if the witness stated that the culprit ‘had a scar on his right cheek’; the length,
orientation and location on the cheek could differ among lineup members. The rationale is
that if the culprit is present, witnesses can use their memory of the distinguishing feature to
identify him. Variation will not bias the lineup against an innocent suspect who the witness
has not seen before. Research on this technique is still in progress. Preliminary data showed
that 40% of witnesses identified a culprit with a scar from a ‘replication with variation’
lineup compared to 25% from an exact ‘replication’ lineup (Valentine and Zarkadi, 2012).
There was no difference in misidentification rates from culprit-absent lineups.
Summary
Face-space, as described in Valentine's (1991a) paper, has inspired many researchers
throughout the world, having been cited more than 600 times (Web of Science). The
implementation of this model has inspired work evaluating theories of caricature and
adaptation effects in face recognition, explaining similarity effects, and the own-ethnicity
Face-space: A unifying concept. 31
bias. In this review, we have described how the face-space has contributed to these areas of
research and also presented new avenues for further research, in which face-space may
provide a unifying explanation for more effects in face recognition. There are clearly many
research questions worth investigating that concern the development of face-space, the neural
representation of face-space, and the application of face-space in forensic and applied
settings. The debate between proponents of exemplar- and prototype-based versions of the
face-space, has been given a new lease of life in the context of facial adaptation and
continues to attract research attention. We hope face-space will continue to inspire new
researchers to explore the fascinating questions of how we process and recognize faces.
References
Allen, H., Brady, N., & Tredoux, C. (2009). Perception of best likeness' to highly familiar
faces of self and friend. Perception, 38, 1821-1830. doi:10.1068/p6424
Anastasi, J. S. & Rhodes, M. G. (2005). An own-age bias in face recognition for children and
older adults. Psychonomic Bulletin and Review, 12, 1043 – 1047. doi:
10.3758/BF03206441
Anastasi, J. S. & Rhodes, M. G. (2006). Evidence for an own-age bias in face recognition.
North American Journal of Psychology, 8, 237 – 253.
Anderson, N. D. & Wilson, H. R. (2005). The nature of synthetic face adaptation. Vision
Research, 45, 1815 – 1828. doi:10.1016/j.visres.2005.01.012
Angeli, A, Davidoff, J. & Valentine, T. (2008). Face familiarity, distinctiveness and
categorical perception. Quarterly Journal of Experimental Psychology, 61, 690-707.
doi: 10.1080/17470210701399305
Balas, B. (2012). Bayesian face recognition and perceptual narrowing in face‐space.
Developmental science, 15, 579-588. doi: 10.1111/j.1467-7687.2012.01154.x
Face-space: A unifying concept. 32
Benson, P. J., & Perrett, D. I. (1991). Perception and recognition of photographic quality
facial caricatures: Implications for the recognition of natural images. European Journal
of Cognitive Psychology, 3, 105-135. doi: 10.1080/09541449108406222
Benton, C. P., & Burgess, E. C. (2008). The direction of measured face aftereffects. Journal
of Vision, 8, 1-6. doi:10.1167/8.15.1
Bernstein, M. J., Young, S. G., & Hugenberg, K. (2007). The cross-category effect: Mere
social categoirzation is sufficient to elicit an own-group bias in face recognition.
Psychological Science, 18, 706-712. doi: 10.1111/j.1467-9280.2007.01964.x
Blakemore, C., Nachmias, J., & Sutton, P. (1970). The perceived spatial frequency shift:
evidence for frequency selective neurones in the human brain. Journal of Physiology,
210, 727 – 750.
Blaney, R. L. & Winograd, E. (1978). Developmental differences in children’s recognition
memory for faces. Developmental Psychology, 14, 441 – 442. doi: 10.1037/0012-
1649.14.4.441
Bruce, V. & Young, A. (1986). Understanding face recognition. British Journal of
Psychology , 77, 305–27. doi: 10.1111/j.2044-8295.1986.tb02199.x
Buckingham G., DeBruine L. M., Little A. C., Welling L. L. M., Conway C. A., Tiddeman B.
P., & Jones B. (2006). Visual adaptation to masculine and feminine faces influences
generalized preferences and perceptions of trustworthiness. Evolution Human Behavior,
27, 381–389. doi:10.1016/j.evolhumbehav.2006.03.001
Burton, N., Jeffery, L., Skinner, A. L., Benton, C. P., & Rhodes, G. (2013). Nine-year-old
children use norm-based coding to visually represent facial expression. Journal of
Experimental Psychology: Human Perception and Performance, 39, 1261-1269. doi:
10.1037/a0031117
Face-space: A unifying concept. 33
Burton, A. M., & Vokey, J. R. (1998). The face-space typicality paradox: Understanding the
face-space metaphor. The Quarterly Journal of Experimental Psychology: Section A,
51, 475-483. doi: 10.1080/713755768
Byatt, G., & Rhodes, G. (1998). Recognition of own-race and other-race caricatures:
Implications for models of face recognition. Vision research, 38, 2455-2468. doi:
10.3758/BF03196628
Carbon C. C., Ditye T. (2012). Face adaptation effects show strong and long-lasting transfer
from lab to more ecological contexts. Frontiers in Psychology, 3, 3. doi:
10.3389/fpsyg.2012.00003
Carbon, C. C. & Leder, H. (2005). Face adaptation: changing stable representations of
familiar faces within minutes? Advances in Experimental Psychology, 1, 1 – 7.
Chance, J. E., Turner, A. L., & Goldstein, A. G. (1982). Development of differential
recognition for own- and other-race faces. The Journal of Psychology, 112, 29 – 37.
doi: 10.1080/00223980.1982.9923531
Cheour, M., Ceponiene, R., Lehtokoski, A., Luuk, A., Allik, J., Alho, K., & Naatanen, R.
(1998). Development of language-specific phoneme representations in the infant brain.
Nature Neuroscience, 1, 351 – 353. doi:10.1038/1561
Chiroro, P. & Valentine, T. (1995). An investigation of the contact hypothesis of the own-
race bias in face recognition. Quarterly Journal of Experimental Psychology, 48A, 879-
894. doi: 10.1080/14640749508401421
Clifford, C. W. G., Wenderoth, P., & Spehar, B. (2000). A functional angle on adaptation in
cortical vision. Proceedings of the Royal Society of London, Series B, 267, 1705 – 1710.
doi: 10.1098/rspb.2000.1198
Face-space: A unifying concept. 34
Craw, I. (1995). A manifold model of face and object recognition. In T. Valentine (Ed.),
Cognitiv e and computational aspects of face recognition. (pp. 183-203.) London:
Routledge.
Davies, G., Ellis, H., & Shepherd, J. (1977). Cue saliency in faces as assessed by the
"Photofit" technique. Perception, 6, 263-9. doi:10.1068/p060263
de Heering, A., Rossion, B., & Maurer, D. (2012). Developmental changes in face
recognition during childhood: Evidence from upright and inverted faces. Cognitive
Development, 27, 17-27. doi: 10.1016/j.cogdev.2011.07.001
Dempster, F. N. (1981). Memory span: sources of individual and developmental differences.
Psychological Bulletin, 89, 63 – 100. doi: 10.1037/0033-2909.89.1.63
Dennett, H. W., McKone, E., Edwards, M., & Susilo, T. (2012). Face aftereffects predict
individual differences in face recognition ability. Psychological Science, 23, 1279-1287.
doi: 10.1177/0956797612446350
Diamond, R. & Carey, S. (1986). Why faces are and are not special: An effect of expertise.
Journal of Experimental Psychology: General, 115, 107-117. doi: 10.1037/0096-
3445.115.2.107
Ellis, H. D. (1975), Recognizing faces. British Journal of Psychology, 66, 409–426. doi:
10.1111/j.2044-8295.1975.tb01477.x
Ellis, H. D., Deregowski, J. B., & Shepherd, J. W. (1975). Descriptions of White and Black
faces by White and Black subjects. International Journal of Psychology, 10, 119 – 123.
doi: 10.1080/00207597508247325
Ellis, H. D. & Flin, R. (1990). Encoding and storage effects in 7-year-olds’ and 10-year-olds’
memory for faces. British Journal of Developmental Psychology, 8, 77 – 92. doi:
10.1111/j.2044-835X.1990.tb00823.x
Face-space: A unifying concept. 35
Feinman, S. & Entwisle, D. R. (1976). Children’s ability to recognize other children’s faces.
Child Development, 47, 506 – 510. doi: 10.2307/1128809
Frowd, C. D. (in press). Facial composite systems. In; Valentine, T. & Davis, J. P. (eds.)
Forensic facial identification: Theory and practice of identification from
eyewitnesses, composites and CCTV. Chicester: Wiley-Blackwell.
Gobet, F. & Simon, H. A. (1996a). Recall of rapidly present random chess positions is a
function of skill. Psychonomic Bulletin and Review, 3, 159 – 163. doi:
10.3758/BF03212414
Gobet, F. & Simon, H. A. (1996b). Templates in chess memory: a mechanism for recalling
several boards. Cognitive Psychology, 31, 1 – 40. doi: 10.1006/cogp.1996.0011
Goldman, D. & Homa, D. (1977). Integrative and metric properties of abstracted information
as a function of category discriminability, instance variability and experience. Journal
of Experimental Psychology: Learning, Memory and Cognition, 3, 375-385. doi:
10.1037/0278-7393.3.4.375
Goldstein, A. G. (1975). Recognition of inverted photographs of faces in children and adults.
Journal of Genetic Psychology, 127, 109 – 123. doi: 10.1080/00221325.1975.10532361
Goldstein, A. G. & Chance, J. E. (1964). Recognition of children’s faces. Child Development,
35, 129 – 136. doi: 10.2307/1126577
Goldstein, A. G. & Chance, J. E. (1980). Memory for faces and schema theory. Journal of
Psychology, 105. 47-59. doi: 10.1080/00223980.1980.9915131
Hancock, P. J., & Little, A. (2011). Adaptation may cause some of the face caricature effect.
Perception, 40, 317-322. doi: 10.1068/p6865
Hills, P. J. (2012). A developmental study of the own-age face recognition bias in children.
Developmental psychology, 48(2), 499-509. doi:10.1037/a0026524
Face-space: A unifying concept. 36
Hills, P. J. (2014). The development of face recognition: Experimental evidence for
maturation. In R. Chen (Ed.), Cognitive Development: Theories, Stages & Processes
and Challenges. Nova-Science Publishers.
Hills, P. J., Holland, A. M., & Lewis, M. B. (2010). Aftereffects for face attributes with
different natural variability: Children are more adaptable than adolescents. Cognitive
Development, 25, 278-289. doi: 10.1016/j.cogdev.2010.01.002
Hills, P. J., & Lewis, M. B. (2006). Reducing the own-race bias in face recognition by
shifting attention. Quarterly Journal of Experimental Psychology, 59, 996-1002. doi:
10.1080/17470210600654750
Hills, P. J., & Lewis, M. B. (2011). Reducing the own-race bias in face recognition by
attentional shift using fixation crosses preceding the lower half of a face. Visual
Cognition, 19, 313-339. doi: 10.1080/13506285.2010.528250
Hills P. J. & Lewis M. B. (2012). FIAEs in famous faces are mediated by type of processing.
Frontiers in Psychology, 3, 256. doi: 10.3389/fpsyg.2012.00256
Hills, P. J., & Pake, J. M. (2013). Eye-tracking the own-race bias in face recognition:
Revealing the perceptual and socio-cognitive mechanisms. Cognition, 129(3), 586-
597. doi: 10.1016.j.cognition.2013.08.012
Hulbert, A. (2001). Trading faces. Nature Neuroscience, 4, 3 – 5. doi: 10.1038/82877
Jaquet, E., & Rhodes, G. (2008). Face aftereffects indicate dissociable, but not distinct,
coding of male and female faces. Journal of Experimental Psychology: Human
Perception and Performance, 34, 101-112. doi: 10.1037/0096-1523.34.1.101
Jeffrey, L. & Rhodes, G. (2011). Insights into the development of face recognition
mechanisms revealed by face after effects. British Journal of Psychology, 102, 799-
815. doi: 10.1111/j.2044-8295.2011.02066.x
Face-space: A unifying concept. 37
Jeffery, L., Rhodes, G., & Busey, T. (2006). View-specific coding of face shape.
Psychological Science, 17, 501-505. doi:10.1111/j.1467-9280.2006.01735.x
Jiang, F., Blanz, V., & O’Toole, A. (2007). The role of familiarity in three-dimensional view-
transferability of face identity adaptation. Vision Research, 47, 525 – 531. doi:
10.1016/j.visres.2006.10.012
Johnson, M. H. & Vecera, S. P. (1996). Cortical differentiation and neurocognitive
development: the parcellation conjecture. Behavioural Processes, 36, 195 – 212. doi:
10.1016/0376-6357(95)00028-3
Johnston, R. A. & Ellis, H. D. (1995). Age effects in the processing of typical and distinctive
faces. Quarterly Journal of Experimental Psychology A: Human Experimental
Psychology, 48, 447-465. doi: 10.1080/14640749508401399
Kaufmann, J. M., & Schweinberger, S. R. (2008). Distortions in the brain? ERP effects of
caricaturing familiar and unfamiliar faces. Brain research, 1228, 177-188. doi:
10.1016/j.brainres.2008.06.092
Knowlton, B. J., & Squire, L. R., (1993).The learning of categories: Parallel brain systems for
item memory and category knowledge Science, 262, 1747-1749. doi:
10.1126/science.8259522
Komes, J., Schweinberger, S. R., & Wiese, H. (2014). Preserved fine-tuning of face
perception and memory: evidence from the own-race bias in high-and low-performing
older adults. Frontiers in aging neuroscience, 6, 60. doi: 10.3389/fnagi.2014.00060
Kuhl, P. K., Williams, K. A., Lacerda, F., Stevens, K. N., & Lindblom, B. (1992). Linguistic
experience alters phonetic perception in infants by 6 months of age. Science, 255, 606-
608. doi: 10.1126/science.1736364
Laurence, S., & Hole, G. (2012). Identity specific adaptation with composite faces. Visual
Cognition, 20, 109-120. doi: 10.1080/13506285.2012.655805
Face-space: A unifying concept. 38
Lee, K. J., & Perrett, D. I. (2000). Manipulation of colour and shape information and its
consequence upon recognition and best-likeness judgments. Perception, 29, 1291-1312.
doi:10.1068/p2792
Leopold, D. A., Bondar, I. V., & Giese, M. A. (2006). Norm-based face encoding by single
neurons in the monkey inferotemporal cortex. Nature, 442, 572-575.
doi:10.1038/nature04951
Leopold, D. A., O’Toole, A. J., Vetter, T., & Blanz, V. (2001). Prototype-referenced shape
encoding revealed by high-level after effects. Nature Neuroscience, 4, 89 – 94.
doi:10.1038/82947
Leopold, D. A., Rhodes, G., Müller, K., & Jeffery, L. (2005). The dynamics of visual
adaptation to faces. Proceedings of the Royal Society B, 272, 897 – 904. doi:
10.1098/rspb.2004.3022
Lewis, M. B. (2004). Face‐space‐R: Towards a unified account of face recognition. Visual
Cognition, 11, 29-69. doi: 10.1080/13506280344000194
Lewis, M. B., & Ellis, H. D. (2000). Satiation in name and face recognition. Memory &
Cognition, 28, 783-788. doi: 10.3758/BF03198413
Lewis, M. B., & Johnston, R. A. (1998). Understanding caricatures of faces. The Quarterly
Journal of Experimental Psychology: Section A, 51, 321-346. doi: 10.1080/713755758
Lewis, M. B. (1999). Are caricatures special? Evidence of peak shift in face recognition.
European Journal of Cognitive Psychology, 11, 105-117. doi:10.1080/713752302
Lewkowicz, D. J. & Ghazanfar, A. A. (2006). The decline of cross-species intersensory
perception in human infants. Proceedings of the National Academy of Sciences of the
United States of America, 103, 6771 – 6774. doi:10.1073/pnas.0602027103
Face-space: A unifying concept. 39
Light, L.L., Kayra-Stuart, F., Hollander, S. (1979). Recognition memory for typical and
unusual faces. Journal of Experimental Psychology: Human Learning and Memory, 5,
212-228. doi: 10.1037/0278-7393.5.3.212
List, J. (1986). Age and schematic differences in the reliability of eye-witness testimony.
Developmental Psychology, 22, 50 – 57. doi: 10.1037/0012-1649.22.1.50
Little, A. C., DeBruine, L. M., Jones, B. C., & Waitt, C. (2008). Category contingent
aftereffects for faces of different races, ages and species. Cognition, 106(3), 1537-
1547. Doi: 10.1016./j.cognition/.2007.06.008
Loffler G., Yourganov G., Wilkinson F., Wilson H. R. (2005). fMRI evidence for the neural
representation of faces. Nature Neuroscience, 8, 1386–1390. doi:10.1038/nn1538
Macchi-Cassia, V., Kuefner, D., Westerlund, A., & Nelson, C. A. (2006). A behavioural and
ERP investigation of 3-month-olds’ face preferences. Neuropsychologia, 44, 2113 –
2125. Doi: 10.1016/j/neuropsychologia.2005.11.014
Mather, G. (1980). The movement aftereffect and a distributed-shift model for coding the
direction of visual movement. Perception, 9, 379 – 392. doi:10.1068/p090379
McLaren, I. P. L. (1997). Categorisation and perceptual learning: an analogue of the
face inversion effect. Quarterly Journal of Experimental Psychology, 50A, 257 – 273.
doi:10.1080/713755705
Medin, D. L. & Schaffer, M. M. (1978). Context theory of classification learning.
Psychological Review, 85, 207-238. doi: 10.1037/0033-295X.85.3.207
Mondloch, C. J., & Thomson, K. (2008). Limitations in 4‐Year‐Old Children’s Sensitivity to
the Spacing Among Facial Features. Child development, 79, 1513-1523. doi:
10.1111/j.1467-8624.2008.01202.x
Moton, J. (1979). Word recognition. In: Morton, J. and Marshall, J. C. (eds.) Psycholinguistic
series, 2: Structures and processes. (pp.107-156.) London: Elek.
Face-space: A unifying concept. 40
Mundy, M. E., Honey, R. C., & Dwyer, D. M. (2007). Simultaneous presentation of similar
stimuli produces perceptual learning in human picture processing. Journal of
Experimental Psychology: Animal Behavior Processes, 33, 124 – 138. doi:
10.1037/0097-7403.33.2.124
Nelson, C. A. (2001). The development and neural bases in face recognition. Infant and Child
Development, 10, 3 – 18. doi: 10.1002/icd.239
Nelson, K. (1996). Language in Cognitive Development. New York: Cambridge University
Press.
Nishimura, M., Maurer, D., & Gao, X. (2009). Exploring children’s face-space: A
multidimensional scaling analysis of the mental representation of facial identity.
Journal of experimental child psychology, 103(3), 355-375. doi:
10.1016/j.jecp.2009.02.005
Nosofsky, R. M. (1986). Attention, similarity, and the identification-categorization
relationship. Journal of Experimental Psychology: General, 115, 39-57. doi:
10.1037/0096-3445.115.1.39
Nosofsky, R. M. (1988). Exemplar-based accounts of relations between classification,
recognition, and typicality. Journal of Experimental Psychology: Learning, Memory,
and Cognition, 14, 700-708. doi: 10.1037/0278-7393.14.4.700
Nosofsky, R. M. (1991). Tests of an exemplar model for relating perceptual classification and
recognition memory. Journal of Experimental Psychology: Human Perception and
Performance, 17, 3-27. doi: 10.1037/0096-1523.17.1.3
Nosofsky, R. M., & Zaki, S. R. (2003). A hybrid-similarity exemplar model for predicting
distinctiveness effects in perceptual old-new recognition. Journal of Experimental
Face-space: A unifying concept. 41
Psychology: Learning, Memory, and Cognition, 29, 1194-1209. doi: 10.1037/0278-
7393.29.6.1194
O’Toole. A. J., Abdi, H. Deffenbacher, K. A. & Valentin, D. (1995). A perceptual learning
theory of the information in faces. In. T. Valentine (ed.) Cognitive and computational
aspects of face recognition: Explorations in face space (pp.159-182). London:
Roultledge.
Over, R. (1971). Comparison of normalization theory and neural enhancement explanation of
negative after effects. Psychological Bulletin, 75, 225 – 243. doi: 10.1037/h0030798
Palmer, S. E. (1975). Visual perception and world knowledge: Notes on a model of sensory-
cognitive interaction. In: D. A. Norman, D. E. Rumelhart & LNR Research Group.
Explorations in cognition. (pp.278-307). San Francisco: W. H. Freeman
Papesh, M. H. & Goldinger, S. D. (2010). A multidimensional scaling analysis of own- and
cross-race face spaces. Cognition, 116, 283-288. doi: 10.1016/j.cognition.2010.05.001
Pascalis, O., de Haan, M., & Nelson, C. A. (2002). Is face processing species-specific during
the first year of life? Science, 296, 1321 – 1323. doi: 10.1126/science.1070223
Perkins, D. (1975). A definition of caricature and caricature and recognition. Studies in the
Anthropology of Visual Communications, 2, 1-24.
Perrett, D. I., May, K. A., & Yoshikawa, S. (1994). Facial shape and judgements of female
attractiveness. Nature, 368, 239-242. doi:10.1038/368239a0
Piaget, J. (1952). The Origins of Intelligence in Children. New York: International
Universities Press.
Pond, S., Kloth, N., McKone, E., Jeffery, L., Irons, J., & Rhodes, G. (2013). Aftereffects
support opponent coding of face gender. Journal of Vision, 13, 16. doi:
10.1167/13.14.16
Face-space: A unifying concept. 42
Pourtois, G., Schwartz, S., Seghier, M. L., Lazeyras, F., & Vullieumier, P. (2005). Portraits
or people? Distinct representations of face identity in the human visual cortex. Journal
of Cognitive Neuroscience, 17, 1043 – 1057. doi: 10.1162/0898929054475181
Reed, S. K. (1972). Pattern recognition and categorization. Cognitive Psychology, 3, 382-407.
doi: 10.1016/0010-0285(72)90014-X
Rhodes, G., Brennan, S., & Carey, S. (1987). Identification and ratings of caricatures:
Implications for mental representations of faces. Cognitive psychology, 19(4), 473-
497. doi: 10.1016/0010-0285(87)90016-8
Rhodes, G. & Jeffery, L. (2006). Adaptive norm-based coding of facial identity. Vision
Research, 46, 2977 – 2987. doi: 10.1016/j.visres.2006.03.002
Rhodes, G., Jeffery, L., Watson, T. L., Clifford, C. W. G., & Nakayama, K. (2003). Fitting
the mind to the world: Face adaptation and attractiveness after effects. Psychological
Science, 14, 558 – 566. doi:10.1046/j.0956-7976.2003.psci_1465.x
Robbins, R., McKone, E., & Edwards, M. (2007). After effects for face attributes with
different natural variability: adapter position effects and neural models. Journal of
Experimental Psychology: Human Perception and Performance, 33, 570 – 592. doi:
10.1037/0096-1523.33.3.570
Rose, S. A., Jankowski, J. J., & Feldman, J. F. (2002). Speed of processing and face
recognition at 7 and 12 months. Infancy, 3(4), 435-455. doi:
10.1207/S15327078IN0304_02
Ross, D. A., Deroche, M., & Palmeri, T. J. (2014). Not just the norm: Exemplar-based
models also predict face aftereffects. Psychonomic Bulletin & Review, 21(1), 47-70.
doi: 10.3758/s13423-013-0449-5
Ross, D. A., Hancock, P. J., & Lewis, M. B. (2010). Changing faces: direction is important.
Visual Cognition, 18(1), 67-81. doi: 10.1080/13506280802536656
Face-space: A unifying concept. 43
Ryu, J. & Chaudhuri, A. (2006). Representations of familiar and unfamiliar faces as revealed
by viewpoint after-effects. Vision Research, 46, 4059 – 4063. doi:
10.1016/j.visres.2006.07.018
Saariluoma, P. (1994). Location coding in chess. Quarterly Journal of Experimental
Psychology, 47A, 607 – 630. doi: 10.1080/14640749408401130
Scott, L. S., Shannon, R. W., & Nelson, C. A. (2005). Behavioral and electrophysiological
evidence of species-specific face processing. Cognitive, Affective, and Behavioral
Neuroscience, 5, 405 – 416. doi: 10.3758/CABN.5.4.405
Scott, L. S., Shannon, R. W., & Nelson, C. A. (2006). Neural correlates of human and
monkey face processing in 9-month-old infants. Infancy, 10, 171 – 186. doi:
10.1207/s15327078in1002_4
Short, L. A., Hatry, A. J., & Mondloch, C. J. (2011). The development of norm-based coding
and race-specific face prototypes: An examination of 5-and 8-year-olds’ face space.
Journal of Experimental Child Psychology, 108(2), 338-357. Doi:
10.1016/j.jecp.2010.07.007
Slone, A. E., Brigham, J. C., & Meissner, C. A. (2000). Social and cognitive factors affecting
the own-race bias in Whites. Basic and Applied Social Psychology, 22, 71 – 84. doi:
10.1207/S15324834BASP2202_1
Solomon, C., Gibson, S. & Maylin, M. (2012) EFIT-V: Evolutionary algorithms and
computer composites. In: Wilkinson, C. & Ryan, C. (eds). Craniofacial identification.
(pp. 24- 41). Cambridge: Cambridge University Press.
Solso, R. & McCarthy, J. (1981). Prototype formation of faces: A case of pseudomemory.
British Journal of Psychology, 72, 499-503. doi: 10.1111/j.2044-8295.1981.tb01779.x
Sporer, S. L. (2001). Recognizing faces of other ethnic groups – an integration of theories.
Psychology, Public Policy, and Law, 7, 36 – 97. doi: 10.1037/1076-8971.7.1.36
Face-space: A unifying concept. 44
Strobach T., Carbon C. C. (2013). Face adaptation effects: reviewing the impact of adapting
information, time, and transfer. Frontiers in Psychology, 4, 318. doi:
10.3389/fpsyg.2013.00318
Tredoux, C. (2002). A direct measure of facial similarity and its relation to human similarity
perceptions. Journal of Experimental Psychology: Applied, 8, 180-193. doi:
10.1037/1076-898X.8.3.180
Tulving, E. and Thompson, D. M (1973). Encoding specificity and retrieval processes in
episodic memory. Psychological Review, 80, 352-373. doi: 10.1037/h0020071
Turk, M. and Pentland, A. (1991). Eigenfaces for recognition. Journal of Cognitive
Neuroscience, 3, 71-86. doi: 10.1162/jocn.1991.3.1.71
Tversky, A. (1977). Features of similarity. Psychological Review, 84, 327-352. doi:
10.1037/0033-295X.84.4.327
Valentine, T. (1986). Encoding processes in face recognition. Unpublished PhD thesis
submitted to the University of Nottingham.
Valentine, T. (1991a). A unified account of the effects of distinctiveness, inversion and race
in face recognition. Quarterly Journal of Experimental Psychology, 43A, 161-204. doi:
10.1080/14640749108400966
Valentine, T. (1991b). Representation and process in face recognition. In: R. Watt (ed.)
Pattern recognition by Man and Machine. (Vol. 14 in 'Vision and Visual Dysfunction'
series edited by J. Cronly-Dillon) (pp. 107-124). London: Macmillan Press
Valentine, T. (2001). Face-space models of face recognition. In: M.J. Wenger & J.T.
Townsend (eds.) Computational, geometric, and process perspectives on facial
cognition: Contexts and challenges (pp. 83-113). Mahwah: LEA.
Valentine, T. & Bruce, V. (1986a). The effects of distinctiveness in recognizing and
classifying faces. Perception, 15, 525-535. doi: 10.1068/p150525
Face-space: A unifying concept. 45
Valentine, T. & Bruce, V. (1986b). Recognizing familiar faces: The role of distinctiveness
and familiarity. Canadian Journal of Psychology, 40, 300-305. doi: 10.1037/h0080101
Valentine, T., Darling, S. & Donnelly, M. (2004). Why are average faces attractive? The
effect of view and averageness on the attractiveness of female faces. Psychonomic
Bulletin & Review, 11, 482-487. doi: 10.3758/BF03196599
Valentine, T. & Endo, M. (1992). Towards an exemplar model of face processing: the effects
of race and distinctiveness. Quarterly Journal of Experimental Psychology, 44A, 671-
703. doi: 10.1080/14640749208401305
Valentine, T & Zarkadi, T. (2012). Testing a new technique for creating lineups for suspects
with distinguishing marks. Poster presented to the 24th Annual Convention of the
Association for Psychological Science, Chicago, USA.
Webster, M. A., Kaping, D., Mizokami, Y., & Duhamel, P. (2004). Adaptation to natural
facial categories. Nature, 428, 557 – 561. doi:10.1038/nature02420
Webster, M. A. & MacLeod, D. I. A. (2011). Visual adaptation and face perception.
Philosophical Transactions of the Royal Society, B, 366, 1702-1725. doi:
10.1098/rstb.2010.0360
Webster, M. A. & MacLin, O. H. (1999). Figural after effects in the perception of faces.
Psychonomic Bulletin & Review, 6, 647 – 653. doi: 10.3758/BF03212974
Wiese H, Komes J, & Schweinberger S. R. (2012). Daily-life contact affects the own-age bias
and neural correlates of face memory in elderly participants. Neuropsychologia, 50,
3496-3508. doi:10.1016/j.neuropsychologia.2012.09.022
Wincenciak, J., Dzhelyova, M., Perrett, D. I., & Barraclough, N. E. (2013). Adaptation to
facial trustworthiness is different in female and male observers. Vision research, 87,
30-34. doi: 10.1016/j.visres.2013.05.007
Face-space: A unifying concept. 46
Winograd, E. (1981). Elaboration and distinctiveness in memory for faces. Journal of
Experimental Psychology, Human Learning and Memory, 7, 181-190. doi:
10.1037/0278-7393.7.3.181
Wright, D. B., & Sladden, B. (2003). An own gender bias and the importance of hair in face
recognition. Acta psychologica, 114, 101-114. doi: 10.1016/S0001-6918(03)00052-0
Yamashita, J. A., Hardy, J. L., De Valois, K. K., & Webster, M. A. (2005). Stimulus
selectivity of figural after effects for faces. Journal of Experimental Psychology:
Human Perception and Performance, 31, 420 – 437. doi: 10.1037/0096-1523.31.3.420
Yin, R. K. (1969). Looking at upside-down faces. Journal of Experimental Psychology, 81,
141-145. doi: 10.1037/h0027474
Zarkadi, T., Wade, K. A., & Stewart, N. (2009). Creating fair lineups for suspects with
distinctive features. Psychological Science, 20, 1448-1453. doi: 10.1111/j.1467-
9280.2009.02463.x
Zhao, L. & Chubb, C. (2001). The size-tuning of the face-distortion after-effect. Vision
Research, 41, 2979-2994. doi: 10.1016/S0042-6989(01)00202-4
Zhao, C., Hancock, P. J., & Bednar, J. A. (2008). Modelling face adaptation aftereffects.
Perception, 37, ECVP Abstract Supplement, 158. doi:10.1068/v080567
Face-space: A unifying concept. 47
Figure captions.
Figure 1: An example of a set of schematic faces. Similar stimulus sets have been used in
many categorization experiments. The faces differ in the distance between the eyes, the
position of the eyes, the length of the nose and the position of the mouth. Reproduced with
permission from Valentine, T. (2001). Face-space models of face recognition. In: M.J.
Wenger & J.T. Townsend (eds.) Computational, geometric, and process perspectives on
facial cognition: Contexts and challenges (pp. 83-113). Mahwah: LEA. © 2001 Lawrence
Erlbaum Associates Inc.
Figure 2: A set of photographs of faces. What are the features on which these faces differ?
Reproduced with permission from Valentine, T. (2001). Face-space models of face
recognition. In: M.J. Wenger & J.T. Townsend (eds.) Computational, geometric, and process
perspectives on facial cognition: Contexts and challenges (pp. 83-113). Mahwah: LEA.
© 2001 Lawrence Erlbaum Associates Inc.
Figure 3: A two-dimensional representation of the face-space as suggested by Valentine
(1991a). Known faces are encoded in a space according to their perceived properties on a
number of dimensions. The two-dimensional form is used for representation only. Adapted
from Valentine, T. (1991a). A unified account of the effects of distinctiveness, inversion and
race in face recognition. Quarterly Journal of Experimental Psychology, 43A, 161-204. doi:
10.1080/14640749108400966.
Figure 4. A two-dimensional representation of the face-space showing populations of own-
ethnicity faces (white circles) and other-ethnicity faces (grey circles). Own-ethnicity faces
show a wider range on the facial dimensions. The two dimensions are for illustration only.
Adapted from Valentine, T. (1991a). A unified account of the effects of distinctiveness,
inversion and race in face recognition. Quarterly Journal of Experimental Psychology, 43A,
161-204. doi: 10.1080/14640749108400966.
Face-space: A unifying concept. 48
Figure 5: The centre images shows a veridical image of Daniel Craig. The image on the right
shows a 30% caricature such that the veridical has been exaggerated away from an average
face by 30%. The image in the left shows a 30% anti-caricature such that the differences
between the veridical and the average face have been reduced by 30%. Reproduced from
Hancock, P. J., & Little, A. (2011). Adaptation may cause some of the face caricature effect.
Perception, 40, 317-322. doi: 10.1068/p6865. © 2011 Pion Ltd London, www.pion.co.uk
and www.envplan.com .
Figure 6: Illustration of adaption to a compressed facial image. Reproduced with permission
from Jeffrey, L. & Rhodes, G. (2011). Insights into the development of face recognition
mechanisms revealed by face after effects. British Journal of Psychology, 102, 799-815.
©2011 The British Psychological Society.
Figure 7: Illustration of the faces, anti-caricatures and anti-faces in face-space. (a) The
original faces (green ellipses) are connected to the average face (blue ellipse) by an ‘identity
trajectory.’ The ‘identity strength’ of the veridical face equals 1.0, and of the mean face
equals 0. Anti-faces on the opposite side of the mean from the original face have negative
identity strengths. (b) Participants viewed a test face for 200 ms and indicated which of four
memorized faces it most closely resembled. Thresholds were reassessed following a 5s period
of adaptation to an anti-face. In the match condition, the test face was on the same identity
trajectory as the anti-face, whereas in the non-match condition it was on a different trajectory.
(c) Three conditions are shown: no adaptation (), adaptation to the matching anti-face ()
and adaptation to the non-matching anti-face (). The y-axis shows the proportion of trials in
which responses matched the identity trajectory of the test face. Adapted with permission
from Macmillan Publishers Ltd: Nature Neuroscience from Leopold, D. A., O’Toole, A. J.,
Vetter, T., & Blanz, V. (2001). Prototype-referenced shape encoding revealed by high-level
after effects. Nature Neuroscience, 4, 89 – 94. doi:10.1038/82947. Copyright 2001. We thank
Face-space: A unifying concept. 49
Chen Zhao for use of his adaptation of the figures [Zhao, C. (2011). Neural Mechanisms of
face and orientation aftereffects. Unpublished PhD thesis. University of Edinburgh].
Figure 8: A model for the neural response to face distortion after-effects. Strength of the
neural response for neural pool A is represented by line A pre-adaptation, and by line A’
post-adaptation, to a face with low eyes. Strength of the neural response for neural pool B is
represented by line B pre-adaptation, and by B’ post-adaptation, to a face with low eyes. The
neural response in pool B is lower following adaptation to an image mainly represented by
this pool, shifting the perceived midpoint from point X to point Y. Copyright © 2007 by the
American Psychological Association. Adapted with permission. The official citation that
should be used in referencing this material is Robbins, R., McKone, E., & Edwards, M.
(2007). After effects for face attributes with different natural variability: adapter position
effects and neural models. Journal of Experimental Psychology: Human Perception and
Performance, 33, 570 – 592. doi: 10.1037/0096-1523.33.3.570. The use of APA information
does not imply endorsement by APA.